Rating: Cautiously Bullish | Target Price: HKD 18.7–25.1 (fair under base case) | Current Price: HKD 18.02 (2026-10-22 close, 52-week low) | Margin of Safety: approx. +4% vs. base-case fair lower bound, probability-weighted expected return +21% | Time Horizon: 12 months | Report Date: 2026-10-23
Previous report: 2026-09-03 (price HKD 22.64, Cautiously Bullish, target HKD 26–32). This is an update report, following the full documentation process.
| Dimension | Previous (2026-09-03, price 22.64) | Current (2026-10-23, price 18.02) | Drivers of Change |
|---|---|---|---|
| Stance | Cautiously Bullish | Maintained | Price decline (-20%) exceeded the downward shift in fair value; relative odds have not deteriorated, but conviction in directional payoff has decreased |
| Confidence | 0.60 | 0.55 (lowered) | Consensus strengthening on 2027 shift to surplus (October Qingdao Aluminum Week collective forecasts by five institutions), Q3 sequential decline, southbound flows turning to selling — macro headwinds to directional payoff have increased |
| Valuation judgment | undervalued | Revised | After the overall downward shift in the fair-value range, the current price is only slightly below the base-case lower bound; evidence for "deep undervaluation" no longer holds |
| Target price range | HKD 26–32 | HKD 18.7–25.1 (down approx. 24%) | FY2027 net profit forecast cut from 29.5–32.5 billion to 24.0–28.5 billion RMB (Q3 -13% QoQ, Antaike 2027 SHFE aluminum at approx. RMB 22,000, LME down 21% from June peak); exit multiple lowered from 7.5–8.25× to 6.5–7.5× (widened peer discount) — both backed by new facts, not merely tracking the stock price |
| Scenario distribution | Bear 14–18@28% / Base 26–32@49% / Bull 36–40@23% | Bear 12.6–15.0@30% / Base 18.7–25.1@48% / Bull 29.6–35.6@22% | Bear case cut: payout-ratio-cut scenario (45–50%) + cycle-bottom PB 0.8–0.95×; expected value 27.4 → HKD 21.8, expected return +21% roughly flat vs. previous (+21.2%), but the distribution has shifted left overall |
Resolution of prior tracking items: The carbon quota policy landed mildly on 2026-09-03 (aluminum smelting bears no substantive carbon cost in 2026–2027; tightening only from 2028) — tail risk eliminated; the USD 330 million 7.05% notes early redemption completed on 2026-09-11 — deleveraging delivered. Buybacks / major shareholder share purchases have not resumed in Sep–Oct (share count flat at 9.819 billion shares for four consecutive months); continue tracking.
China Hongqiao is the world's second-largest electrolytic aluminum producer (2025 output of 6.545 million tonnes, 14.8% of China's total). Its ore–alumina–power–aluminum–deep-processing integration gives it a fully costed cost per tonne (estimated at approx. RMB 13,030) about 18% below the industry average. The stock is down 47% YTD, closing at HKD 18.02 on October 22 for a 52-week low, driven by a triple overlay: LME aluminum down 21% from the June geopolitical peak to USD 3,006; five institutions collectively forecasting a global shift to surplus in 2027 (+410,000 to +1.4 million tonnes); and beta from the fading dividend-style rally in HK stocks. The current price implies PB of approx. 1.15× (0th percentile over the past year), FY2026E P/E of approx. 4.4–4.7×, and FY2025 dividend yield of 9.2%. If the 60% payout ratio persists, FY2026E DPS would be approx. HKD 2.30–2.48, a yield of 12.8%–13.7%. This report judges the current price as "fair but cheap": the market has already priced in the deep 2027 surplus and a payout-ratio cut, while FY2026 net profit of RMB 32.0–34.5 billion is essentially locked in, and the balance sheet (net gearing approx. 15%, debt/EBITDA 1.1×, cash and equivalents RMB 52.4 billion) supports continued high payouts. Base-case fair value is HKD 18.7–25.1 (+22% to midpoint), probability-weighted expected value HKD 21.8 (+21%); the main risk is overseas capacity landing on schedule in 2027 pushing SHFE aluminum below RMB 20,500 (bear case HKD 12.6–15.0, -23%). The final dividend declared with the March 2027 annual results is the validation point for the entire thesis: a final dividend ≥220 HK cents triggers the dividend re-anchoring; below 165 HK cents, this judgment is falsified.
What has changed: ① FY2026 earnings uplift is essentially locked in — H1 net profit RMB 17.21 billion (+39.2%); A-share subsidiary Hongqiao Holdings' first-three-quarters net profit RMB 23.29 billion (+69.9%); the full-year path to RMB 32.0–34.5 billion is clear, but Q3 alone was approx. RMB 7.70 billion, -12.8% QoQ, with momentum falling from its peak; ② the 2027 balance-sheet expectation has collectively shifted from shortage (2026 deficit of 710,000–1.6 million tonnes) to surplus (Goldman +1.30–1.50 million / CRU +1.40 million / Morgan Stanley +840,000 / UBS +900,000 / Macquarie +410,000 tonnes, with only Antaike maintaining a tight balance), with the dispersion across institutions exceeding 2 million tonnes — this is the entire core of the current valuation dispute; ③ the stock is down 20% since 9/2, of which roughly two-thirds is HK beta (Hang Seng down 9.5% from its 9/9 high), and southbound funds have turned to net selling over the past 5 days.
Primary driver: the SHFE aluminum price path (every ±RMB 1,000/tonne ≈ net profit ∓/±3.5–4.0 billion RMB). Leading indicators: LME inventory (236,000 tonnes, lowest since 1990), domestic social inventory (560,000–640,000 tonnes in mid-to-late October, a historical low for the period), and the 2027 pace of supply additions in Indonesia / the Middle East / India (Indonesia's planned 2027 additions of approx. 3.435 million tonnes; Middle East restart from 4.8 to 6.4 million tonnes).
Expectation gap: The market is effectively pricing a 45–50% payout scenario (demanding an 11%–13% yield on FY2026E dividends), while the balance sheet and two consecutive years of actual 62% payout support continuation at 60%; on the sell side there are zero sell ratings and an average target price of HKD 33.8–35.3 (+87% to +96% vs. current price) — the dispute lies mainly between sell-side and the market. Verification catalysts: positive profit alerts in Jan–Feb 2027, the final dividend in the March 2027 annual results, and share-count changes in the monthly returns (buyback resumption). Falsification conditions: final dividend <165 HK cents; SHFE aluminum monthly average persistently <RMB 20,500; Indonesian capacity ≥3 million tonnes commissioned in 2027 with inventory restocking.
Multi-period evidence of earnings quality is solid: OCF/net profit for FY2023–2025 of 1.95/1.52/1.72 respectively, and 1.29 in 2026H1; FCF/net profit for FY2023–2025 of 1.43/0.96/1.25; estimated ROIC of 14.6%–16.7%. Recurring net profit in 2026H1 was approx. RMB 17.46 billion, with non-recurring items (fair-value gain on CB derivative instruments RMB 0.86 billion, plant impairment -0.58 billion, inventory write-down -0.53 billion) totaling only approx. 1.4% of net profit — high earnings quality. But two points must be candidly stated: first, the fully costed cost per tonne of approx. RMB 13,030 is derived in this report by backing out ASP minus gross profit per tonne (flat for two consecutive years), while industry fully costed (tax-inclusive) costs have converged toward the company as alumina prices collapsed (-17.8% YoY) — in a falling alumina environment, the integration advantage manifests as "giveback of external alumina sales profit"; it is a two-way amplifier, not a one-way moat. Second, half of H1's approx. RMB 18.4 billion free cash flow increase came from capex contraction (-62.6%), i.e., an investment-cycle trough; the period end still carried contracted capital commitments of RMB 4.80 billion and Simandou funding obligations of up to USD 1.78 billion, plus RMB 10.2 billion of zero-coupon CBs maturing in May 2027 — the "high cash content" conclusion holds on a multi-year average basis and should not be linearly extrapolated from H1's FCF level. Q3 (approx. RMB 7.70 billion on the Hongqiao Holdings basis, down approx. 12.8% from Q2's RMB 8.83 billion, though still +55.6% YoY) confirms earnings momentum fading from its peak, mainly due to lower external alumina selling prices and sequentially weaker aluminum prices.
The correct reading of the low multiples: TTM earnings are at a historical extreme (the industry's September profit per tonne of RMB 8,382 is an all-time high on Baichuan Yingfu spot data), so the low PE percentile is a cyclical-peak illusion and serves only as context; the primary anchor should be PB×ROE and dividends. PB of 1.15× is a 36% discount to the A-share peer median (1.81×), while TTM ROE of approx. 21% is above the peer average — part of that discount reflects the A/H structural premium and cyclical phase difference; the cleaner comparable is Chalco H (PE approx. 7.3×), against which Hongqiao remains notably cheap. Three layers of value: asset floor approx. HKD 19.6–21.2 (BVPS 15.76 + capacity-quota scarcity premium of HKD 3.87–5.41/share); zero-growth EPV HKD 21.5–29.6 (normalized net profit RMB 22.0–26.0 billion, FCFE conversion 0.90–0.95, cost of equity 10.5%) — both independent references exceed the current price. What the market is effectively pricing: a 9.2% yield on the FY2025 dividend of HKD 1.65 and 11%–13% on FY2026E dividends — i.e., treating high payouts as untrustworthy peak cash flow. A caveat on basis: all per-share metrics use 9.819 billion shares without potential dilution — the USD 300 million CB due 2030 (conversion price HKD 18.44, only 2.3% above the current price, potential +1.3% share count) and Hongqiao Holdings' RMB 12 billion private placement (group holding 88.99%→80.90%, approx. 8% net profit dilution) together drag per-share metrics by approx. 4%–9%; backing out "market-implied normalized earnings" from selected multiples is assumption-dependent and serves only as supporting evidence.
The track record is consistent: FY2023–2025 DPS of 63/161/165 HK cents, with payout ratios of approx. 62% for both FY2024–2025; buybacks in 2025 + 2026H1 totaling HKD 10.87 billion, all cancelled; the major shareholder bought 41.5 million shares in June (average price approx. HKD 26.4, now approx. 32% under water). But three facts constrain the "committed high-dividend" narrative: first, 2026H1 dividends + buybacks totaled RMB 18.7 billion = 109% of the period's net profit, relying on existing cash and unsustainable; second, buybacks have been suspended since June, share count in the monthly return has been flat for four consecutive months, and the major shareholder made no new purchases in Sep–Oct — four months with no insider buying around HKD 18 (the USD 147 million buyback earmark contractually arranged at the 2026 CB issuance must be deployed before 2026-12-31, a Q4 watch item); third, the 2027 cash schedule is dense: RMB 10.2 billion of zero-coupon CBs maturing 2027-05-04 (conversion price HKD 43.90, a 144% premium to the current price — conversion hopeless, must be cash-settled) + final dividend of approx. RMB 19.0–21.0 billion + Simandou funding commitments, with net debt likely to rebound from RMB 21.3 billion. This report uses a 60% payout ratio as the base-case assumption (momentum of two consecutive years of actual 62% + balance-sheet support): FY2026E DPS approx. HKD 2.30–2.48, yield 12.8%–13.7%; the conservative 50% case implies HKD 1.91–2.06, 10.6%–11.5%. If the payout ratio falls below 45%, the dividend valuation anchor fails and pricing shifts to cycle-bottom PB — the March 2027 declaration is the sole validation point.
The attribution of the price decline must be precise: LME's 21% fall from the June peak of USD 3,787.5 was mainly the unwinding of the war premium from the Feb–June US–Iran conflict after the memorandum of understanding; over the same period SHFE aluminum fell only approx. 6% — a 3.4× gap between the onshore and offshore declines; the unwinding of the geopolitical premium does not imply an equivalent deterioration in fundamentals. But the evidence of cyclical position still points to a high level: the industry profit per tonne of RMB 8,382 is a historical extreme (spot basis), and Hongqiao's H1 realized gross profit per tonne of RMB 8,162 is likewise near its listing high. The 2027 balance sheet is the entire dispute: five institutions forecast a surplus of 410,000–1.5 million tonnes (dispersion exceeding 2 million tonnes), while Antaike views it merely as inventory normalization from the 2026 deficit of 710,000–1.6 million tonnes; on the sell side, no single broker has yet cut earnings on the surplus thesis (Morgan Stanley raised its target to HKD 29.7 on 10-21; Soochow raised 2028E net profit to RMB 34.8 billion in mid-October) — the transmission from "balance-sheet surplus" to "earnings downgrades" has not yet appeared in sell-side forecasts, and that is exactly where the bull-bear standoff lies. This report's sensitivity analysis (baseline: FY2026E RMB 32.0–34.5 billion, 2026 realized SHFE aluminum average approx. RMB 23,300; each ±RMB 1,000/tonne ≈ net profit ∓3.5–4.0 billion, plus alumina net-long linkage and impairment effects): 2027 SHFE average RMB 22,000 → net profit approx. RMB 24.0–28.5 billion; RMB 20,500 → approx. RMB 19.0–22.0 billion (DPS approx. HKD 1.2–1.5); RMB 19,000 → approx. RMB 13.0–16.5 billion (DPS approx. HKD 0.7–1.0). The latter two correspond to fair value of HKD 12.6–15.0 (cycle-bottom PB 0.8–0.95×).
The claim that "the 45-million-tonne red line welds supply shut" must be corrected: the red line constrains compliant new capacity, not actual output — installed capacity already slightly exceeds the red line, and 1–5% over-production headroom exists under high profits, which is why Goldman raised its China output forecast. The real price safety cushion lies in inventory (LME at 236,000 tonnes, lowest since 1990; domestic social inventory at a historical low for the period) and the fact that within 2026 only 300,000–500,000 tonnes of high-certainty overseas additions are due. But in 2027, overseas supply (Indonesia's 3.435 million tonnes planned, Middle East restart +1.6 million tonnes, India's 4.8 million tonnes planned) will take over marginal pricing, and such plans have historically often been delayed — the pace of commissioning is itself the largest scenario variable. The company-level structural buffers (Yunnan green-power migration, Indonesian alumina expansion, deep processing with H1 volumes +23.2% / gross margin 32.7%) can improve cost and product mix, but cannot hedge a downward shift in the aluminum price center; moreover, the company runs a net-long alumina position (FY2025 output 13.397 million tonnes vs. own electrolytic aluminum demand of approx. 12.5 million tonnes), and with a global alumina surplus of approx. 3.7 million tonnes in 2026 and industry losses of approx. RMB 112/tonne, the alumina segment (H1 gross margin 6.3%) is a structural drag. Also note: under current EU CBAM rules only direct emissions count, so the international compliance value of a "green-aluminum export premium" is currently zero; green-power benefits mainly accrue via domestic carbon constraints and power-price structure.
| Metric | FY2023 | FY2024 | FY2025 | 2026H1 | First Three Quarters 2026 (Hongqiao Holdings basis) |
|---|---|---|---|---|---|
| Revenue (RMB 100mn) | 1,336.2 | 1,561.7 | 1,623.5 | 875.1 (+8.0%) | 1,316.6 (+13.7%) |
| Net profit attributable to parent (RMB 100mn) | 114.6 | 223.7 | 226.4 | 172.1 (+39.2%) | 232.9 (+69.9%) |
| Recurring net profit (RMB 100mn, this report's basis) | — | — | 264.2 | 174.6 | 231.1 (excluding non-recurring items) |
| Gross margin | 15.7% | 27.0% | 25.6% | 31.5% | — |
| Net margin attributable to parent | 8.6% | 14.3% | 13.9% | 19.7% | 17.7% |
| Operating cash flow (RMB 100mn) | 224.0 | 339.8 | 390.0 | 221.5 | 263.6 |
| Free cash flow (RMB 100mn, OCF − capex) | 163.6 | 213.7 | 283.4 | 184.5 | — |
| Cash + cash-like assets (RMB 100mn) | — | 447.7 | 511.9* | 524.1* | — |
| Interest-bearing debt (RMB 100mn) | — | 729.9 | 884.3* | 674.4 | — |
| Debt-to-asset ratio | 47.0% | 48.2% | 42.2% | 40.5% | — |
| Debt/EBITDA | — | — | 1.8× (2025H1) | 1.1× | — |
| Dividend per share (HK cents) | 63 | 161 | 165 | No interim dividend | — |
Note: *Cash-like assets/interest-bearing debt at end-2025 and 2026H1 are period-end balance sheet stock figures (2026H1 cash-like assets = cash 461.4 + time deposits 35.2 + restricted deposits 27.5 RMB 100mn; interest-bearing debt includes guaranteed notes of RMB 4.06bn and the derivative-instrument component of convertible bonds of RMB 11.85bn); first three quarters 2026 figures are for the A-share subsidiary Hongqiao Holdings (002379, Group holding 88.99%) on a basis not consolidated with minority interests, provided for trend reference only.
Drivers of Metric Changes (YoY ≥ ±20% summary): 2026H1 net profit attributable to parent +39.2% — attributed by the company to higher aluminum alloy product prices (ASP +18.7%; LME +31.9% YoY, SHFE aluminum +20.7%); alumina segment revenue -22.1% / gross margin -22.5pp — attributed to falling alumina prices (ASP -28.3%); capex -62.6% — decline from the investment cycle of major projects such as the Yunnan park; finance costs -13.6% — debt reduction and lower financing rates; deep-processing 2025 gross margin -6.7pp — lower capacity utilization and cancellation of export tax rebates. FY2024 net profit attributable +95.2% — the super-cycle in alumina (segment gross margin 35.4%) resonating with rising aluminum prices. First three quarters Hongqiao Holdings +69.9% — a sharp YoY rise in aluminum prices; Q3 standalone -12.8% QoQ — falling external alumina selling prices and weaker QoQ aluminum prices (the voluntary announcement offered no attribution; this is the market consensus interpretation based on segment volume/price data).
On October 22, the same day as the Q3 report, the share price fell -4.76% to a 52-week low (HSI -3.03% that day; roughly two-thirds was market beta). The numbers themselves were not bad: Hongqiao Holdings' first-three-quarters revenue RMB 131.66bn (+13.7%), net profit attributable RMB 23.29bn (+69.9%), non-recurring profit RMB 23.11bn, operating cash flow RMB 26.36bn (+43.3%) — consistent with Huachuang's FY2026E group path of RMB 33.67bn. What the market read was the sequential trend: Q1 RMB 6.76bn → Q2 RMB 8.83bn → Q3 RMB 7.70bn (estimated), -12.8% QoQ — alumina futures at RMB 2,628/ton in October (nearly halved from 5,000+ a year earlier in 2025) eroding profits on external sales beyond the integrated chain, and LME aluminum -7.5% since end-September. Extrapolating on the H1 group/subsidiary profit ratio (~1.10), group Q3 profit was roughly RMB 8.4bn and first three quarters roughly RMB 25.6bn; assuming SHFE aluminum at RMB 23,000–23,700/ton in Q4, full-year net profit attributable would be RMB 32–34.5bn. Absent access to consensus sell-side estimates, this report benchmarks the above range against individual sell-side forecasts (Huachuang 33.67 / Guosheng 35.3 RMB bn): the midpoint of RMB 33.3bn is about 1% below Huachuang. The verdict on "beat or miss" awaits the profit alert in January–February 2027, when the expected final dividend (≥220 HK cents) will be the key variable for the stock.
Business model: An asset-heavy, fully vertically integrated commodity business — Guinean bauxite (held via a 22.5% associate interest in the SMB-Winning Consortium, not consolidated) → alumina (~19 million tonnes domestic capacity plus 2 million tonnes in Indonesia, more than self-sufficient) → captive power / Yunnan hydropower → electrolytic aluminum (6.46 million tonnes of compliant capacity) → deep processing / recycled aluminum. Revenue is 100% product sales priced at prevailing market rates — no recurring revenue, no pricing power. Differentiation lies in cost (integration + captive power) and green-aluminum premiums (management says downstream customers pay a low-carbon premium for green aluminum).
Cash content of earnings: OCF/net profit attributable FY2023–2025 = 1.95/1.52/1.72 (three-year average 1.73); FCF/net profit attributable = 1.43/0.96/1.25 — book profits have consistently converted fully into cash, with no bloated receivables buildup (receivables turnover ~20 days). On a recurring basis: FY2025 net profit attributable of RMB 22.64bn vs. RMB 26.42bn after excluding convertible bond derivative losses (derivatives were -RMB 3.78bn in 2025 due to mark-to-market swings on zero-coupon CBs, reversing to +RMB 0.86bn in 2026H1) — consecutive zero-coupon CB issuances over two years have made MTM gains/losses swing by as much as ±15% of net profit attributable, so this noise must be stripped out when reading the P&L (the non-GAAP adjustments are cross-verified by company disclosures and our own calculations, not aggressive add-backs). One point to watch: in 2026H1 receivables grew +11.6%, faster than revenue (+8.0%); the company itself noted "extended payment terms for some customers," with the 3–12 month aging bucket rising to 59.5% — a second consecutive year of loosening terms, a mild yellow flag rather than a red one.
Return on capital: Our estimated ROIC for FY2024/2025 was ~16.7%/14.6%; the company's disclosed return on capital (average equity) for 2025 was 18.6% — above the 10%–11% cost of equity, a genuine moat range. Even in the earnings downcycle (FY2023) the company earned RMB 11.46bn, proving its ability to traverse cycles from the left side of the cost curve.
Capex quality: FY2024 RMB 12.6bn → FY2025 RMB 10.7bn → 2026H1 RMB 3.7bn — the investment peak has passed and the company has entered harvest mode. The company does not disclose a maintenance/growth capex split (maintenance spend cannot be determined), but the allocation toward the Yunnan relocation + lightweighting + new energy aligns with capital commitments of RMB 4.8bn, directed at green power substitution (lower carbon and lower power-price risk). Red flag: alumina segment gross margin collapsed from 35.4% to 6.3% over two reporting periods — cyclical stalling of a single profit engine rather than a sign of fraud; H1 dividends + buybacks = 109% of net profit attributable, meaning the pace of returns relies on existing cash.
Consistency of words and deeds: pragmatic. ① "Continue to reduce debt" → bank borrowings -17.4% in 2026H1, Debt/EBITDA from 1.8× to 1.1×, and in September a further redemption of USD 330mn of 7.05% notes (delivered). ② "Ongoing buybacks based on intrinsic value" → HKD 5.58bn in 2025 + HKD 5.286bn in 2026H1, all cancelled; the controlling shareholder added 41.5mn shares in June (delivered, though suspended after June). ③ The 2024 annual report's "seize opportunities in bauxite and alumina prices" → the 2025–2026 alumina windfall was given back (returned after the cycle peaked — beyond management's control). No habit of quantitative guidance, no record of exaggerated forecasts; all outlooks are qualitative.
Shareholder friendliness: neutral-to-friendly, but not committed. DPS rose from 63 → 161 → 165 HK cents, with payout ratios of ~62% for two consecutive years; HKD 10.87bn of buybacks cancelled over two years. Offsetting items: in November 2025 a "top-up placement" of 400mn shares (HKD 29.20) raised net proceeds of HKD 11.49bn — the largest refinancing since listing; net share count rose ~+4.8% in 2025 (later partially offset by buybacks); no written dividend policy; the RMB 10.2bn zero-coupon CB was cleverly timed (issued at a 25.3% premium), but its derivative P&L persistently distorts the income statement.
Risk signals: The Chairman also serves as CEO (deviation from the CG Code disclosed); family trusts hold 62.58%; large volumes of related-party transactions, all disclosed per rules (bauxite procurement, the RMB 240mn photovoltaic acquisition from Weiqiao Xinchuang, etc.; the 20.28% appraisal premium is supported by market comparables); on 2027-01-13, ~465mn shares of Hongqiao Holdings held by financial investors will be unlocked (3.6%–4.3% of total share capital) — the nearest certain supply event. Fragments circulating online about "executives selling 4.01mn shares" have not been verified via exchange disclosures of dealing in shares and are not credited in this report.
| Segment | Revenue share (2026H1) | Gross margin | Revenue YoY | Business logic in one line |
|---|---|---|---|---|
| Aluminum alloy products (electrolytic aluminum) | 68.1% | 38.5% (FY2025: 28.5%) | +14.8% | Captive power + integration cost advantage; greatest profit elasticity as spreads widen |
| Alumina | 18.4% | 6.3% (FY2025: 22.2%; FY2024: 35.4%) | -22.1% | Full rollercoaster of the price cycle; the 2024 windfall fully returned |
| Aluminum alloy deep processing | 11.9% | 32.7% (FY2025: 19.2%) | +39.7% | Volume and price both rising; export tax rebate shock absorbed by price recovery |
Main profit driver: Of group gross profit of ~RMB 27.35bn in 2026H1, aluminum alloys contributed RMB 22.94bn (83.9%), deep processing RMB 3.40bn (12.4%), and alumina only RMB 1.01bn (3.7%) — the alumina segment, the second largest by revenue, is now virtually profitless; group earnings are effectively a single-variable function of "electrolytic aluminum spread × 6.46 million tonnes of capacity." The gross margin gap (38.5% vs 6.3%, a 32pp spread) reflects two entirely different businesses: electrolytic aluminum earns scarcity rents under the capacity red line, while alumina is in a global capacity-clearing phase with a 3.7-million-tonne surplus — integration is a two-way amplifier during alumina downturns (cost savings on captive use, losses on external sales). This structural feature makes group earnings more sensitive to aluminum prices than pure smelting peers.
Accounting red flags: ① Twice-loosened receivables terms (medium) — in both 2024 and 2026H1 the company noted "extended payment terms for some customers"; receivables growth outpaced revenue, with the 3–12 month aging bucket rising to 59.5%; management's explanations were consistent. ② Convertible bond derivative P&L distortion (low) — fair-value changes of derivatives from 2023 to 2026H1: -0.5 → -21.9 → -37.8 → +0.95 RMB 100mn, amplifying in sync with CB issuances; the company has not attributed these separately (a mechanical result of zero-coupon CB accounting, not manipulation, but it pollutes the net-profit-attributable figure). ③ Inventory write-downs and reversals within the same period (low) — H1: write-downs of RMB 530mn and reversals of RMB 680mn; net reversals lowered current-period COGS, with amounts swinging with metal prices.
Cross-period consistency: Across the full alumina cycle of segment gross margins 11.1% → 35.4% → 22.2% → 6.3%, the company attributed each period's results to prices, with direction and magnitude matching the numbers (consistent); the derivatives line item was unexplained by the company (inconsistent — unexplained); wording on receivables-term expansion was consistent across two years (consistent). Overall judgment: based on the five reporting periods sampled here, FY2023–2026H1, no obvious financial-engineering traces found — goodwill/intangibles ~0.1% of assets, receivables turnover 20 days, OCF persistently above profit; cross-period anomalies concentrate in two explainable factors, CB accounting and cyclical prices.
Reserves and output: The company does not disclose reserve data (HK-listing disclosures cover sales volumes only). Sales volumes: electrolytic aluminum FY2025 5.824 million tonnes (flat) / 2026H1 2.811 million tonnes (-3.3%); alumina 13.397 (+22.7%) / 6.917 million tonnes (+8.6%); deep processing 716 (flat) / 444 thousand tonnes (+23.2%). Bauxite is held via a 22.5% associate interest (SMB-Winning shipped 71.52 million tonnes in 2025, expected above 90 million tonnes in 2026; theoretical attributable self-sufficiency ~70%); the Group's bauxite procurement budget for 2026 is ~USD 3bn; reserve replacement ratio not applicable (associate basis).
Unit economics: Estimated fully-costed cash cost ~RMB 13,030/tonne (back-calculated from ASP minus gross profit per tonne, flat for two consecutive years), ~18% below the industry's fully-loaded cost including tax of RMB 15,866–16,100/tonne, placing it in the top quartile of the global cost curve (benchmarked against industry averages; not verified plant-by-plant against the original CRU curve). Gross profit per tonne was ~RMB 8,162 in 2026H1 (FY2024/2025 ~RMB 4,317/5,183).
Hedging and sensitivity: No aluminum/alumina commodity hedges disclosed (FX management only; FX losses of RMB 420mn in H1) — earnings are fully exposed to commodity prices. Sensitivity: each ±RMB 1,000/tonne move in SHFE aluminum ≈ ∓RMB 3.5–4bn in net profit attributable (~±12%); each -RMB 100/tonne in external alumina selling prices ≈ -RMB 300–500mn in net profit.
Geopolitics and mining rights: Guinea's 2026 export quota cap of 150 million tonnes (-18% vs 2025) has been postponed three times without taking effect; export tariffs have already been raised from 5.5% to 10%; the government mandates that foreign investors build alumina refineries locally or forfeit mining rights; an escalation toward a bauxite export ban is a tail risk (estimated annualized impact on net profit attributable of -RMB 3–4bn, probability ~5%–8%); Simandou iron ore equity commitments capped at USD 1.78bn (a further ~USD 600mn still to be invested), with the associate's profit -41% in H1 (early-stage cost drag).
NAV perspective: The asset floor is ~HKD 19.6–21.2/share = BVPS 15.76 + scarcity premium on capacity quotas of HKD 3.87–5.41 (6.46 million tonnes × RMB 5,000–7,000/tonne, anchored to Nanshan Aluminum's 2023 transaction precedent of RMB 6,633/tonne). One popular narrative must be honestly corrected: EV/capacity ≈ RMB 26,600/tonne, far above the electrolysis segment's replacement cost of RMB 4,500–5,500/tonne — "market cap below replacement cost" does not hold. The NAV-discount argument should be precisely framed as "the scarcity premium on capacity quotas is not fully priced" (floor midpoint HKD 20.4 vs current price 18.02).
Current market data (2026-10-22 close): Share price HK$18.02 (52-week low, YTD -47%), market cap HK$176.95bn; PE-TTM ~5.5× (0% 1-year / 12% 3-year / 35% 5-year percentile—TTM earnings in an extreme zone, percentile for reference only); FY2025 static 6.7×; FY2026E 4.4–4.7×; PB 1.15× (0% 1-year / 43% 3-year / 66% 5-year percentile, 5-year median 0.95×); FY2025 DPS yield 9.2%.
| Peer | PE-TTM | PB | ROE (period) | Key differences |
|---|---|---|---|---|
| China Hongqiao | ~5.5× | 1.15× | ~21% (TTM) | Lowest-cost position + ~62% payout ratio, H-share discount |
| Chalco A/H | 8.6× (A) / ~7.3× (H) | 1.77× | 19.9% (TTM) | SOE full-chain, largest scale, 40% payout ratio |
| Yunnan Aluminum | 7.8× | 2.24× | 21.9% (H1) | Purest hydro green aluminum, 15.8% debt ratio |
| Shenhuo | 7.0× | 1.84× | 27.3% (first three quarters) | Coal-aluminum dual business, dual cyclic exposure |
| Tianshan Aluminum | 7.4× | 1.59× | 13.5% (H1) | Xinjiang captive-power integration, one tier smaller |
(Peers are A-share basis, around 2026-10-22; ROE periods differ, for horizontal reference only.)
Market-implied expectations: At the current price, the FY2025 dividend implies a 9.2% required yield and the FY2026E dividend (HK$2.30–2.48) implies 12.8%–13.7%—the market treats the high payout as untrustworthy peak cash flow; back-solving at cycle-midpoint multiples implies normalized net profit of ~RMB 19.0–23.0bn (below FY2025 actual of RMB 22.6bn, 30%+ below sell-side FY2027E; back-solving depends on multiple assumptions, supplementary only). Against reality: FY2026 net profit of RMB 32.0–34.5bn is largely locked in, and the 2027 dispute is between RMB 24.0–28.5bn (Antaike path) and sell-side RMB 33.7bn. In one sentence: the current price requires earnings to halve in 2027 with payout cut below 50%, a combination unsupported by the balance sheet and two years of payout momentum—the discount is backed by the 2027 surplus consensus, not pure mispricing, but its magnitude is excessive.
Three-layer value (EPV): Asset floor HK$19.6–21.2 (independent reference); zero-growth EPV HK$21.5–29.6 (normalized net profit RMB 22.0–26.0bn, FCFE conversion 0.90–0.95, Ke 10.5%, midpoint HK$25.3)—current price is ~29% below the EPV midpoint, i.e., the market pays nothing for growth and applies a near-30% discount for cycle credibility; growth options account for ~-41% of the current price (this report's calculation), with Yunnan green power/Indonesia/deep-processing upgrades valued at zero. The asset floor and EPV are both above the current price while the stock makes new lows, indicating marginal pricers are bidding on the bear case (impairments + PB 0.8×).
Three scenarios and odds (per-share fair ranges, current price 18.02):
| Scenario | Probability | Fair range | vs. current | Key drivers |
|---|---|---|---|---|
| Bear: 2027 surplus arrives early | 30% | 12.6–15.0 (mid 13.8) | -23% | SHFE aluminum 2027 average <RMB 20,500/t, net profit RMB 13.0–22.0bn, DPS HK$0.7–1.5, payout cut to 45–50%, cycle-bottom PB 0.8–0.95× |
| Base: Antaike path | 48% | 18.7–25.1 (mid 21.9) | +22% | SHFE aluminum 2027 average RMB 21,500–22,500/t, FY2027 net profit RMB 24.0–28.0bn × 6.5–7.5× (anchor: A-share peer TTM PE median 7.6× with H-share discount tier, Chalco H at 7.3×, dividends HK$2.0–2.3 cross-checked against 9%–11% required yield) |
| Bull: shortage persists + dividend re-anchoring | 22% | 29.6–35.6 (mid 32.6) | +81% | SHFE holds above RMB 23,500/t, Indonesia/Middle East supply delayed, FY2026–27 net profit RMB 33.0–35.0bn × 7.5–8.5×, DPS ≥HK$2.6 (range consistent with sell-side target cluster of HK$29–33.8) |
Probability-weighted expectation HK$21.8 (+21%), margin of safety ~+4% vs. base-case fair lower bound (thin), odds structure moderately positive (+81% / -23%). Bear-case anchor check: lowest sell-side target of HK$26 (Goldman Sachs, 2026-06 Neutral) is still far above this bear case—no bears on the broker side; this bear case takes the valuation implications of the 2027 surplus path (net profit RMB 13.0–22.0bn + cycle-bottom PB + payout cut, triple overlay), more conservative than Goldman's single-stock target; the gap mainly reflects this report's incorporation of a dividend-logic reset from 62% payout to 45–50%, whereas all sell-side targets assume an unchanged payout.
In-house earnings forecast and reconciliation anchors:
| Period | Revenue (RMB bn) | Net profit attributable (RMB bn) | Key assumptions |
|---|---|---|---|
| FY2026E | 175.0–179.0 | 32.0–34.5 (mid 33.3) | H1 actual 17.21; Q3 ~8.4 (estimated); Q4 SHFE aluminum RMB 23,000–23,700/t, volume ~5.8mn t, unit cost flat at RMB 13,030/t, drag from external alumina sales |
| FY2027E | 168.0–176.0 | 24.0–28.5 | SHFE average RMB 21,500–22,500/t (Antaike ~RMB 22,000/t path), partial Indonesia/Middle East supply, A-share placement dilution not included |
Comparison: Huachuang (10-14 maintained) FY2026E/2027E RMB 33.67/33.76bn, Guosheng (August) 35.3/33.9bn—this report's FY2026 aligns with sell-side; FY2027 is 15%–29% lower, with the entire gap in 2027 aluminum prices (this report assigns weight to the surplus path). Methodology note: Q3 group figure extrapolated at 1.10× Hongqiao Holdings' A-share figure (H1 empirical ratio); once the placement completes, this coefficient becomes invalid and must be re-estimated.
Conclusion: reasonably undervalued. Assess quality (left of cost curve, ROIC ~15%, cash conversion 1.7×, 62% payout) and price (PB at 0% 1-year percentile, FY2026E forward yield 12%+) separately: quality is excellent while price has partly priced in the cyclical downturn—not the deep undervaluation of early September ("current price at an 80% discount to mid-cycle earnings"), but a left-side zone where "the bear case has real probability and the base case offers a thin margin of safety." Strategic implication: phased accumulation beats lump-sum; the window before the final dividend declaration in March 2027 is the key validation point.
Industry size and supply-demand: 2025 global electrolytic aluminum output 74.23mn t (Antaike basis, of which China ~44.15mn t, nearly 60%); at LME US$3,000/t, metal value ~US$220bn-class; Jan–Aug 2026 China output 31.12mn t (+3.9%, NBS), IAI global output growth down to within 1.2% for Aug–Sep. China's supply story has ended institutionally: 45mn t capacity ceiling (compliant cap ~45.43mn t), as of end-June 2026 built capacity 45.17–46.10mn t, operating 44.67–45.31mn t, utilization 97.7%–99.7%—but the "welded shut" narrative must be corrected: at high profits, restarts and over-production carry 1–5% flexibility; Goldman Sachs has raised its China 2027 output forecast to 46.3mn t (1.3mn t above the red line), and Antaike expects China 2027 output growth <1%. Global increments lie overseas: 2027 planned overseas additions ~3.435mn t (Indonesia 8 projects, 3.51mn t planned, 65.8%; India ~4.8mn t planned concentrated around 2027; Middle East 2.39–2.71mn t restarts under disruption, CRU expects regional output 480→640mn... wait, 4.8mn t 2026 → 6.4mn t 2027); 2027–2028 overseas supply growth 11.5%/7.9% (CSC/东方财富 research basis). The balance sheet is the industry-wide pricing core: 2026 global deficit (CRU -710k t, Wood Mackenzie -900k t, Morgan Stanley -1.1mn t, Emkay -1.6mn t, wide dispersion); 2027 institutions collectively shift to surplus—Goldman +1.3–1.5mn t (2027 LME average US$2,750), CRU +1.4mn t, UBS +900k t, Morgan Stanley +840k t (2027 LME could fall to US$2,800), Macquarie +410k t (2027 LME US$3,050), Antaike tight balance (2027 domestic average ~RMB 22,000, 2026 RMB 24,100–24,200)—the spread exceeds 2mn t, meaning "whether 2027 is deficit normalization or a supply glut" remains unresolved, while inventories (LME 236k t, lowest since 1990; domestic social inventory 560–640k t, historical seasonal low) temporarily side with tight balance. No platform-level demand inflection: China real domestic demand -0.6% YoY in 2026H1 (Mysteel ex-exports), apparent consumption propped by exports (2026 aluminum product exports expected >10mn t vs. 9.2mn t in 2025, Antaike); NEV/PV increments fully priced; the "AI data-center aluminum" narrative lacks volume (2026 ~809k t → 2030 ~1.2mn t, Morgan Stanley, only ~1.1% of global demand—IAI-cited commentary calls it "still a rather small share"); property completions and PV aluminum demand under pressure in 2027 (Antaike).
Value chain and value distribution: Bauxite (Guinea-dominated, 2025 China imports 149mn t, ~74% of imports; October Guinea ore CIF US$72/dry t, nearly halved from early-2025 peak of US$114) → Alumina (the worst link in the chain currently: 2026 global surplus ~3.7mn t, spot RMB 2,654.6/t, -53.4% from late-2024 peak of RMB 5,700; September industry full cost incl. tax RMB 2,789.57/t, industry-wide loss ~RMB 112/t; Indonesia adding ~3mn t capacity in 2026, historically turning net exporter) → Electrolytic aluminum (profit concentration point: September spot settlement RMB 24,248/t, industry full cost incl. tax RMB 15,866, theoretical profit RMB 8,382/t at a historical extreme—essentially a profit redistribution of "alumina falling to a cost floor while aluminum holds high") → Processing/exports (European duty-paid premium ~US$520/t supports export arbitrage). Hongqiao spans the full chain: alumina self-sufficiency with surplus (net long position), ore locked via 22.5% SMB-Winning equity—falling alumina is a cost windfall for pure smelters but a small net loss for Hongqiao after internal hedging (external sales hurt, internal use benefits; group-level impact discounted to 60–70%); the electrolytic aluminum segment retains the industry's highest gross margin.
Supply-demand and competitive landscape: None of the demand drivers (property completions, NEV lightweighting, PV frames, UHV/grid, exports) has pricing power; prices follow the cycle. On supply, China is nominally "frozen" into an oligopoly—top five players ~41.9% of national output (2025): Chalco 8.08mn t/18.3% first, Hongqiao 6.545mn t/14.8% second, Yunnan Aluminum ~7%, Shenhuo ~4%, Tianshan ~3%. Entry barriers = scarce capacity-swap quotas (policy red line) + capital intensity + power and bauxite endowments; domestic competition is locked into a cost competition, with the real frontline shifting overseas in 2027–2028 (Indonesia/India new entrants + Middle East restarts competing for increments). Substitution threat: recycled aluminum (near-zero CBAM surrender obligations; 2024–2025 China output target 11.5mn t, still small). Bargaining power: strong vs. downstream (spot premiums, export premia), weak vs. upstream bauxite (Guinea 74% of imports with frequent policy disruptions)—integrated players hedge via equity-locked resources.
Cycle and regulation: Price at cycle mid-point (LME down 21% from June geopolitical peak of US$3,787.5 to US$3,006, mostly war-premium unwind; SHFE down only ~6% from highs to RMB 23,100–23,700); earnings at a historical extreme percentile (profit expansion from "premium falling, costs falling to floor"). Leading indicators: LME/social inventories, SHFE near-month backwardation (near tight, far loose), Middle East restart progress, Indonesia ramp-up, China monthly exports. Four regulatory lines: ① capacity red line (institutional moat, soft constraint); ② carbon market—on 2026-09-03 MEE issued the 2025/2026 quota plan; aluminum smelting bears no substantive carbon cost in 2026–2027 (at current carbon prices, quota cost ~RMB 2.6/t), tightening in 2028; green power consumption ratio requirement raised to 30%+ by 2027 (three years earlier than the original 2030, long-term raising coal-power aluminum costs, favoring Hongqiao's Yunnan hydro + captive-power retrofit); ③ export VAT rebate cancellation (13% rebate removed from 2024-12, offset by overseas premia; 2026 exports still expected >10mn t); ④ EU CBAM effective 2026-01 (direct emissions only: recycled aluminum near-zero cost, green-power aluminum's paper advantage temporarily zero, primary aluminum default-value path ~€157/t—Hongqiao's long-term compliance cost for coal-power aluminum exports to Europe, no actual payment in 2026).
Company positioning: World's second-largest and largest private electrolytic aluminum producer; its 6.46mn t domestic quota is an unreplicable license asset under the capped landscape; moat = the world's most complete integrated chain (Guinea equity ore theoretical self-sufficiency ~70% + alumina self-sufficiency + captive power/Yunnan hydro), unit costs ~18% below industry average, top quartile of the global cost curve (estimate, to be verified). Stable share; excess profit comes from "cost-side left" rather than demand positioning; structural upgrades (Yunnan green power migration plans of 2.176/3.96mn t, Indonesian alumina, deep-processing ramp) provide a medium-term buffer. Market position basis note: the 14.8% domestic output share/world No.2 rankings follow Antaike and other third-party bases; the company's annual report calls itself "a leading global aluminum product manufacturer"—the two are consistent, with no basis conflict.
Judgment: Cautiously bullish, 12-month horizon. The current price of HK$18.02 (2026-10-22 close) has pre-priced "deep 2027 surplus + payout cut," while FY2026 earnings are locked, the balance sheet supports sustaining a 60% payout, and the asset floor (~HK$20) and zero-growth EPV (~HK$25) both sit above the current price—base-case fair value HK$18.7–25.1 (+22% to midpoint), probability-weighted expectation HK$21.8 (+21%), downside to bear-case midpoint -23%. Expected return is flat versus the prior report six weeks ago, but the distribution has shifted left and the validation window is nearer. The odds structure at this level suits phased accumulation rather than a concentrated bet: the thin margin of safety (only +4% vs. the base-case lower bound) means a 2027 surplus realization still leaves ~20% downside; dividends (9.2% on FY2025 paid basis, 12.8%–13.7% FY2026E forward) are the main left-side buffer.
Strategy points: ① Validation before adding—positive profit alert in Jan–Feb 2027 and the final dividend declaration in March (≥HK22.0 cents confirms dividend re-anchoring; <HK16.5 cents falsifies, exit) are decision points; ② aluminum price thresholds—between the RMB 22,000/t tier (base) and RMB 20,500/t tier (bear) of SHFE monthly averages is the add/trim boundary; ③ watch the Q4 buyback earmark (US$147mn, must be executed by 2026-12-31) and share-count changes in monthly reports; a resumed buyback is the hardest bottom signal; ④ the 2027-01-13 unlock of ~465mn Hongqiao Holdings shares and the RMB 10.2bn CB redemption on 2027-05-04 are two certain calendar events; watch their impact on A/H sentiment and the cash schedule.
Risk warnings: Overseas supply landing on schedule in 2027 pushing SHFE below RMB 20,500/t (earnings and dividend double-hit); payout cut to 45–50% (dividend valuation anchor fails, switching to cycle-bottom PB); Guinea quota/ban-style escalation (tail risk -RMB 3.0–4.0bn/yr); beta drag from a sustained fade in the HK dividend style; potential dilution from the A-share placement and CB (combined 4%–9%); alumina segment turning to losses resonating with impairments. Should any of these materially occur, this report's valuation framework must be rebuilt.